[R-SIG-Finance] stochastic oscillator OBOS - intraday data & optimization TIMEFILTER & TIMESPANS

[amarjit chandhial]

Hi Ilya,


Have returned with our stochastic oscillator OBOS strategy.

Where we were last time was optimizing nSlowD & the OBOS thresholds. I will guess that everything was correct there, inequalities etc, if not let me know! Therefore I'll carry-on with including a timefilter to our strategy and optimizing timespans (various timefilters). As this is for demo and given only a few days of intraday data the values for nSlowD and OBOS thresholds can remain at 5, 80, & 20, respectively (ordinarily this'll be on the optimized values). I have also uncommented the short-side as we have an fx rate GBPUSD, not a stock.


Attached:
(1) Vanilla + inclusion of a timefilter, chosen arbitrarily as 'T07:00/T20:00'. This appears to be filtering correctly.
(2)  Various timefilters i.e. timespans for optimization. Although the program runs, it does not appear to be optimizing. I am having problems with add.distribution and apply.paramset functions. Please help!
 


Amarjit 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20140516/07b2c5b1/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ac_stochOSC_OBOS_GBPUSD_vanilla_TIMEFILTER.R
Type: application/octet-stream
Size: 7604 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20140516/07b2c5b1/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ac_stochOSC_OBOS_GBPUSD_vanilla_TIMESPAN.R
Type: application/octet-stream
Size: 3735 bytes
Desc: not available
URL: <https://stat.ethz.ch/pipermail/r-sig-finance/attachments/20140516/07b2c5b1/attachment-0001.obj>